SMART GRID LOAD

FORCASTING (short term)

USING BIG DATA ANALYSIS
M.TECH (1ST SEMESTER) Developers

 MAJOR : EDPS  Presentation by :

Tuhin Shubra Das
 Department Of Electrical (EDPS, 1st year , 1st sem)
Engineering
 Guided By:
 JIS Collage of Engineering , Mr. Rajat Bhattacharya
Kalyani, West Bengal Dr. Papan Biswas
 Referred documents & data source
 Zeyar Aung, Mohamed Toukhy (Masdar Institute of Science and Technology) & John R. Williams, Abel Sanchez, Sergio
Herrero (Massachusetts Institute of Technology) : Towards Accurate Electricity Load Forecasting in Smart Grids
 Mohamed H. Toukhy; Data Mining Techniques for Smart Grid Load Forecasting
 Vikram Gandotra : Smart grids concepts ; SIEMENS Smart grid division
 COLUMBIA UNIVERSITY: http://www.ldeo.columbia.edu/res/pi/4d4/testbeds/Smart-Grid-White-Paper.pdf
 An introduction to smart grid ; US Energy department
 B. R. More ; Electric Load Forecasting in Smart Grid environment and Classification of Methods
 Sneha Vasudevan : One step ahead in Smart grid Load forcasting, Ohio State University
 Tao Hong,: Energy Forecasting in the Smart Grid Era: Opportunities, Challenges and Case Studies; SAS Institute
 Kurt Reisacher: Totally integrated IT/OT framework for next generation DMS ; Siemens Energy Management – Smart
Grid Solutions
 Dr. Jinghan He: Short-term load forecasting based on big data technology; School of Electrical Engineering Beijing
Jiaotong University
 Ron Jarvis; SAP for utilities, Siemens
 Steve Pascoe; Big Data Analytics: Maximizing Enterprise Data; EMC 2016 Technology Conference
 Energy-Efficient Infrastructure Management in the IoT era; George Dritsanos, IT BU VP for Greece, Cyprus & Malta ;
Schneider Electric
 Mohamed Eltabakh : Large-Scale Data Management & Big data Technology presentation
 Ruoming Jin ‘Xinyu Chang : Introduction to Big Data ; http://www.cs.kent.edu/~jin/BigData/index.html
 Hafedh Yahmadi : Internet of Things A presentation
 Dr.-Ing. Abdur Rahim ; Empowering Urban Innovation Through Convergence of IoT-Big data and Cloud; Forum on
Internet of Things: Empowering the New Urban Agenda Geneva, Switzerland, 19 October 2015
 Dr Payam Barnaghi, Dr Chuan H Foh; Internet of Things A presentation and lecture; University of Surrey
 agenda
Concept of SMART GRID
Concept of INTERNET OF THINGS
Concept of CLOUD COMPUTING
Concept of BIG DATA TECHNOLOGY
Need of SHORT TERM LOAD FORCASTING in SMART
SYSTEMS
Drawbacks of existing techniques
Proposed HADOOP (BIG DATA) Application
Analysis of results
 Concept of smart grid
 There are many views of what is In reality, a smart grid is not a single concept but rather a combination of
technologies and methods intended to modernize the existing grid in order to improve flexibility,
availability, energy efficiency, and costs.

A smart grid is an electricity network that uses digital and other advanced technologies to monitor and manage
the transport of electricity from all generation sources to meet the varying electricity demands of end users.

Smart grids co-ordinate the needs and capabilities of all generators, grid operators, end-users and electricity
market stakeholders to operate all parts of the system as efficiently as possible, minimizing costs and
environmental impacts while maximizing system reliability, resilience and stability.

Smart grids are a set of evolving technologies that will be deployed at different rates in variety of settings
depending on local conditions such as existing technologies, regulatory frameworks and investment
framework.
 smart grid attributes
Information-based
Communicating
Secure
Self-healing
Reliable
Flexible
Cost-effective
Dynamically controllable
Elements of a Smart Grid
RELIABILITY AND MANAGED OPERATIONAL
RESOURCE OPTIMIZATION
EFFICIENCY PLANNING RELIABILITY

Load Development (profile & season)

Use of micro/distributed generation Reconfiguration for reliability

Advantageous control methods Strategic switching placement

and strategies
Reliability contribution
Auxiliary device control and Equipment:
Virtual Generation Planning measurement •Condition Monitoring
•Alarming
Anticipative reconfiguration for: Local generation management •Forecast Loading vs. Capability
•Contingencies •Condition Assessment
•Losses Catastrophic weather planning on-the-fly: •Crew Work Order Generation
•Voltage Profile •Local Area Severity Anticipation •JIT Service Kit Inventory
•Micro Generation Utilization •Reconfiguration
•Short Circuit Duty •Reconnection Asset Location/Relocation Based on
•Crew Management and Dispatch Field Experience and Measurement
On-the-fly:
•Contingency Planning Automated outage:
•Reconfiguration •Identification and Logging
•Losses •Reconfiguration based on Last Status
•Voltage Profiling •Local Generation Dispatch
•Micro Generation Scheduling

Smart Grid
YesterdaY’s and todaY’s Grid
Before

 One-way limited communication  Few sensors and analog control

 One-way power flow  Little to no consumer choice
 No electric vehicles  Reactive maintenance
 Centralized generation  Limited usage transparency
Tomorrow's Grid
 Bi-directional communication  Pervasive monitoring and digital control
and metering
 Self-monitoring & high visibility
 Bi-directional power flow
 Many consumer choices
 Millions of electric vehicles
 Condition-based maintenance
 Applications
 Proliferation of numerous applications

After

Power

Information
 Internet of things
 From any time ,any place connectivity for anyone, we will now have
connectivity for anything!
10
the Model has ChanGed…
 The Model of Generating/Consuming Data has
Changed

Old Model: Few companies are generating data, all others are consuming data

New Model: all of us are generating data, and all of us are consuming data
 Convergence in EVERYWHER
 Convergence of technology
 Convergence of business
and ecosystem
 Convergence of people,
application, things, data,
devices, etc.
 Convergence of Electrical
Grid & Power Systems

HYBRID
Convergence of Internet of
things (IoT), big data and cloud
 For IoT, number of billions of connected devices is an indicator
of IoT. The connectivity is just an enabler but the real value
of IoT is on data (Technical & business insight/data-driven
economy)
 For Big Data, data collection is one of the main concern, and
IoT can play an important roles for data collection and data
sharing
 For Big Data, data is nothing without real Techno-Commercial
business value insight
 Cloud offers Everything as a Service business model for IOT
and big data.
 IoT is a King, Big data is a Queen and Cloud
is a Palace
Key requirements of
IoT-Big data-cloud platform
Security and
privacy

Intelligent and
Scalable dynamic

Distributed and
Real-time decentralized
What Is Cloud Computing
 Generally speaking, cloud computing can be thought of as anything that involves
delivering hosted services over the Internet.

 According to National Institute of Standards and Technology (NIST), USA,

Cloud computing is a model for enabling ubiquitous, convenient, on-demand
network access to a shared pool of configurable computing resources (e.g.,
networks, servers, storage, applications, and services) that can be rapidly
provisioned and released with minimal management effort or service
provider interaction. (Special Publication 800-145)

It is a means of:
outsourced shared-computing where resources
are virtualised, distributed and pooled amongst external data centres
accessed by users through the internet
 Cloud : application
Cloud computing has been credited with increasing competitiveness through cost
reduction, greater flexibility, elasticity and optimal resource utilization. Here are a few
situations where cloud computing is used to enhance the ability to achieve Technical and
business goals.
1. Infrastructure as a service (IaaS) &
platform as a service (PaaS)
2. Private cloud and hybrid cloud
3. Test and development of Mathematical
Models (Server less computing)
4. Big data analytics
5. File storage
6. Disaster recovery
7. Backup
Big Data Definition
 No single standard definition…
“Big Data” is data whose scale, diversity, and complexity require new
architecture, techniques, algorithms, and analytics to manage it and
extract value and hidden knowledge from it…

20
Characteristics of Big Data:
1-Scale (Volume)
 Data Volume
 44x increase from 2009 2020
 From 0.8 zetta bytes to 35zb
 Data volume is increasing exponentially

Exponential increase in
collected/generated data

21
Characteristics of Big Data:
2-Complexity (Variety)
 Various formats, types, and
structures
 Text, numerical, images, audio,
video, sequences, time series, social
media data, multi-dim arrays, etc…
 Static data vs. streaming data
 A single application can be
generating/collecting many types
of data

To extract knowledge all these types of

data need to linked together

22
Characteristics of Big Data:
3-Speed (Velocity)
 Data is begin generated fast and need to be processed
fast
 Online Data Analytics
 Late decisions  missing opportunities
 Examples
 E-Promotions: Based on your current location, your purchase history,
what you like  send promotions right now for store next to you
 Healthcare monitoring: sensors monitoring your activities and body
 any abnormal measurements require immediate reaction
 Smart Grid (micro grid) Monitoring & Control: sensors monitoring
grid can do SCHEDULING of Microgrid for particular AREA according
to the SHORT TERM LOAD FORCASTING PROFILE of the given Area

23
BiG data: 3V’s

24
soMe Make it 4V’s

25
 Who’s GeneratinG BiG data

Mobile devices
(tracking all objects all the time)

Social media and networks Scientific instruments

(all of us are generating data) (collecting all sorts of data)

Sensor technology and

networks
(measuring all kinds of data)
 The progress and innovation is no longer hindered by the ability to collect data
 But, by the ability to manage, analyze, summarize, visualize, and discover knowledge
from the collected data in a timely manner and in a scalable fashion

26
27
Harnessing Big Data

 OLTP: Online Transaction Processing (DBMSs)

 OLAP: Online Analytical Processing (Data Warehousing)
 RTAP: Real-Time Analytics Processing (Big Data Architecture & technology)

28
Big Data Technology

29
Short term load forcasting
 Short term load forecasting refers to forecasting load curve
of coming several hours ,one day or several days ahead. It is
not only ensure power system to operate safely and
economically, and also lay the foundation of formulating
electricity dispatching schedule and transactions.

Short-term load forecasting accuracy has been listed as an

important assessment indicator in Grid Corporation. With the
development of power market, the requirement of forecasting
accuracy, timeliness, reliability and intelligence become higher
load forecasting algorithm
Problems of algorithm
NEW APPROACH / RESEARCH NEEDED
Steps of calculation (without BIG
DATA technology : hadoop )
 STEP 1 : Improved Hierarchical Cluster Technique
Classify daily load curves of each user into several types using improved hierarchical
cluster technique.

Daily load curve

Hierarchical Cluster Result
Step 2:Grey Relation Analysis
Step 3: establishing the decision
tree
Forecasting Total System Load
 The total system load is forecasted based on aggregation of individual load’s
forecasting results as shown in Figure below :
 Repeat the above steps for each user, then aggregate all of them.

The forecasted total load can be calculated

by adding all of the forecasted individual
load together with line loss .

Distribution
Transformer

Feeder
Implementing Hadoop Platform
(big data analysis Technology)
 The former steps are just one distribution transformer’s forecasting process. To get REAL TIME system load
forecasting result. We need to repeat this for all distribution transformers and then add them up by big
data technology
Error % : Hadoop vs SVM MODEL
 The traditional approach which use SVM model and only focus on system load has
 Max relative error: 5.20% Min relative error: 0.39% Average relative error: 1.61%
 It means the BIG DATA ANALYSIS is better than traditional approach.

STUDIED Dis. Transformer Load TOTAL System load

TRADITIONAL: HADOOP : ERROR %

ERROR %
questions